mariovitali
Senior Member (Voting Rights)
Martha-Spyridoula Katsarou , Maria Papasavva , Annia Tsolakou , Avgi Christodoulou, Archontoula Antonoglou , Athanasios Raptis , Antonios Kontaxakis , Marios Gavrielatos, Ioannis Michalopoulos, Alice G. Vassiliou , Maria Stefanatou, Georgios Pappas , Sterghios A. Moschos l, Nikolaos Drakoulis , Paraskevi Katsaounou
Abstract
Although the COVID-19 pandemic has now been down-graded, long COVID (LC) presents an ongoing risk of long-term disease for a significant percentage of the population, even after mild or no symptoms upon infection. LC post-viral effects have been associated with oxidative stress (OS), impacting canonical cell function. The aim of this study was to investigate the association of eight OS-related single nucleotide polymorphisms (SNPs) on LC susceptibility among patients with mild or no symptoms during SARS-CoV-2 infection, with emphasis on a clinically homogeneous population free from bias and overlap with other conditions.
Blood samples were collected from 85 clinically confirmed LC patients and 96 unvaccinated controls (observational case control study) all with mild/asymptomatic infection, and analysed by targeted SNP genotyping in the GSTP1, SELENOS, CAT, SOD2, and EPHX1 OS-related genes. Τhe control individuals had been infected at least 6 months prior to enrollment and had not developed any symptoms related to long COVID. Our analysis revealed associations between SOD2 and EPHX1 polymorphisms and disease progression, with pre-existing thyroid disease and acute phase symptoms being significant aggravating factors. Machine Learning (ML) analysis produced a 10-factor predictive model for LC with a balanced accuracy over 0.74, released herein as an open-access LC risk rating webtool.
Our findings suggest that individuals’ genetic antioxidant capacity may plays an important role in long covid, fitting with current ideas of mitochondrial dysfunction and viral persistence. It is also shown how well diagnosed and bias free cohorts can reveal patterns often missed in self-reported cases and the potential for predictive tools that combine genetic and clinical data.
Web | DOI
Last edited by a moderator: